Optimization of Control Line Encapsulation Based on Numerical

Simulations of Shock and Vibration

Luiz Fernando Bermero Nardi1; Daniel Fraga Sias2; Leonardo Teixeira Fernandes Abreu2; Mariana Silva
Guimarães2; Manoel Feliciano Da Silva1; Thiago Handerson Torres Eduardo1
1
PETROBRAS – Petróleo Brasileiro S.A.
2
WIKKI BRASIL – Consultoria em Engenharia LTDA

Paper presented at the Offshore Technology Conference, Virtual and Houston, Texas, August 2021.
Paper Number: OTC-31222-MS

https://doi.org/10.4043/31222-MS

Published: August 09 2021

Abstract

Control line integrity is fundamental to ensure operability of intelligent completion systems. Damage to
control lines can normally occur during string deployment (by shock), or during productive life (by
vibration-related fatigue).This work aimed to investigate how encapsulation protection will affect
control line survivability throughout their entire life cycle.
Numerical simulations were carried out using geometrical arrangements of control lines typically found
in completion valves and tubings. For installation loads, lines were submitted to shock simulations, both
inside and outside of protection clamps. The role of flatpack material was studied, as to which kind of
material would better absorb impact energy.For productive life loads, the flow-induced vibration was
taken into account for high-rate injector wells. Vibration loads were basedonforce results obtained by
Computational Fluid Dynamics (CFD). Structural Finite ElementAnalysis (FEA) was performed to
obtain local tensions to estimate productive life of lines.

Introduction
Intelligent completion systems use control lines to provide power (hydraulic) and signals (electric)
to/from downhole tools. Damage to lines can happen during installation of the completion string as
described by (Mitchel, 2010) and (Rodrigues, 2017). Impact of unprotected control lines against the
casing wall can lead to crushing or denting of such lines. In case of denting, line survivability is
expected to decrease, as predicted by the works of(Liman, 2012)and(Cosham, 2004).
Another source of damage is flow-induced vibration of control lines (FSI), since lines are exposed to
highly turbulent annular flows on injection wells. According to (Nardi, 2018), Reynolds numbers in the
order of10 can be expected. Vibration effects can be more severeif flow characteristical frequency
matches the natural frequenciesof structures, leading to resonance. Less intense vibration can
nevertheless cause fatigue-related damage, especially on long-term operation.
Vibration mitigation on tubular structures has been targeted by many researchers, such as(Hahn,
Masson, Norris, & Williams, 2015)and(Azizian & Torrado, 2016) .Vibration mitigation usually involves
changes to the geometry of the problem, with the addition of joints and couplings. Ideal mitigation can
be accomplished byconstraining the vessel structure at multiple points. An example of this approach is
presented by (Aggogeri, Borboni, Merlo, Pellegrini, & Ricatto, 2017), who employed foam
2

encapsulations surrounding the entire outer surface of vibrating pipes. This type of protection is not
always applicable, since system cavities are not always capable of accommodating anti-vibration
devices.
If structural reinforcement cannot be added to the system, improved material selection rises as one
important option to deal with vibration. Viscoelastic materials are especially capable of absorbing
vibration, and have been used throughout many industries, to this end. A complete review of viscoelastic
damping materials can be found on (Zhou, Yu, Shao, Zhang, & Wang, 2016).
(Geethamma, Asaletha, Kalarikkal, & Thomas, 2014)explains the three different types of damping that
can be provided by viscoelastic materials, according to three molecular mechanisms involved: coulomb
damping, viscous damping and hysteretic damping. Hysteretic damping produces heat by internal
friction of the polymer chain is the damping mechanism of most interest to the present work.
Also according to (Geethamma, Asaletha, Kalarikkal, & Thomas, 2014), there are two types of
dampeners, depending on the type of response: passive dampeners and active dampeners. This work will
only approach passive dampeners, which consist of inertial components added to the structure. Active
vibration control can be achieved with reverse piezoelectric materials and can be even more efficient
than passive control.
According to (Mohammadi & Sedeghati, 2012)slipping can happen between elastic and protective
viscoelastic materials. Slipping effects reduce damping properties and should be avoided. (Bai & Sun,
1995) cited by (Mohammadi & Sedeghati, 2012)obtained controversial results for damping in some
materials. It was obtained that dampingefficiency can be different for each vibration mode of a given
structure.

Recommended practices for encapsulation design

Control lines are manufactured using a metallic tube, which can be seamless or seam-welded. Typically,
the metallic material specification is N08825 or S32750, according to the Unified Numbering System
(UNS) standard. Material selection, for metal tubes, is based on collapse and burst requirements and
corrosion resistance.
Encapsulation materials are typically polymers such as FEP (Fluorinated Ethylene Propylene), ETFE
(Ethylene Tetrafluoroethylene), or PVDF (Polyvinylidene fluoride) depending on chemical
compatibility and temperature requiriemnets of the application scenario.
Encapsulation is responsible for protecting lines against shockand abrasion during installation in the
well. Three recommended practices from AWES (Advanced Well Equipment Standards): (3362-AWES-
FP-MAIN-FINAL, 2017), (3362-AWES-CL-MAIN-FINAL, 2017),and (3362-AWES-PR-ENCAP,
2020)are important references for qualification tests to prevent failures during deployment due to reel
and skid handling, control line handling, maximum tesile strength, encapsulation slip, clamp interfaces,
and connectors interfaces. Despite such efforts, there is a lack of recommended practices to prevent
failures as consequence of vibration during well production and injection.
The main source of vibration is the flow through the intelligent completion valve due to the high
frequencies and forces involved. It is important to note that assembling done onshore and on the RIG
can lead to denting and microfractures, which can accelerate the failure (initial damage). Denting can be
prevented by implementing quality control on transportation and assembling of completion tools,
including the use of special handling equipment.

Failure mechanisms
The following failure mechanisms were approached in this work:
 3

• Vibration damage - vibration

ibration can
c cause resonance, flutter, and fatigue, depending on the natural
modes of vibration of the structure and on the natural frequencies of structure and surrounding
flow. Dry connector/control
nector/control line interfaces are especially subjected to fatigue.
• Fretting damage- will
ill happen especially in the presence of metal-to-metal
metal metal friction and is related
to wear; Equipment/control
quipment/control line interfaces are especially subjected to fretting.
• Shock damage – occurs when material yields or ruptures, due to the kinetic energy absorbed
absorbe by
shock.

Methodology

Shock simulations
ines were impacted against casing walls, with typical trip-in
Encapsulated lines trip velocities. Heave and
sway velocities were considered to account for potential failure in drill string compensator. Impact
velocities were as low as 2.16ft/s.
/s. Impacting mass was chosen as the typical mass of a completion valve,
180 lb.Several
.Several configurations of contro
controll line were studied, including materials and pre-tension
pre states.

Vibration simulations
While most of the references in literature focus in minimizing transmissibility coefficient, this work
presents a new methodology to address vibration issues by calculating the fatigue damage associated
with long-term
term vibration on structures. Despite being more costly, this method allows for greater
precision on the calculation of damage to lines, and is able to point out which locations are more
susceptible for damage. Figure 1 shows a schematic of the several steps adopted for this calculation.

Figure 1 - Roadmap employed for coupled CFD –FEA analyses

4

The first step of vibration analyses was to obtain pressure and velocity fields of the flow surrounding
control lines in the annular space. Due to the high turbulence of this flow, Large L Eddy Simulations
(LES) were chosen. Line proximity to the jet nozzles showed to play an important role on the force llevel
experienced by these lines. In the event that lines are poorly affixed onto the valve body, they risk
slipping towards position 1, ass depicted in Figure 2 (a), where side forces will be far more intense.

Figure 2–Set-up for CFD analysis of flow around injection valve; (a) Transversal section at flow ports, showing different
dif positions
for control lines; (b) 3D-view
view of valve, with force distribution over one of the lines.

Once the pressure field was obtained, surface integrals were numerically performed over the control
lines surfaces, as shown in Figure 2 (b). Finally, a transient structural analysiss was conducted with these
loads to obtain tension history on the control lines as a function of time, as shown in Figure 3.

Figure 3 – Stress concentration on free line

Since tension observed in control lines was below 10% of their yield stress, S-N
S methodology was
chosen to evaluate fatigue. According to (Miner, 1945), the accumulated damage can be calculated by a
linear method, as per eq. (1).

1
1
 5

is number of cycles with amplitude and is number of cycles with respect to fatigue failure in the
same stress amplitude.If the stress amplitude( is a continuous function, the
t damage can be be
given by eq. (2).

is the load cycles frequency

quency and T is the duration period until failure(Hahn,
(Hahn, Masson, Norris, &
Williams, 2015).In the present study, fatigue was calculated by using the cycle
cycle-counting algorithm
Rainflow. This method counts cycles of a random load ssignal. The Palmgren-Miner
Palmgren rule was also
applied to account for different amplitudes.

Results

Shock
Figure 4 shows the tension response of control lines for a shock againt the production casing wall.
Three different
ent encapsulation materials were employed: PVDF (Polyvinylidene
Polyvinylidene fluoride),
fluoride PP
(Polypropylene) and nylon.. Impacts lasted for about 20 ms,, but most of the fluctuation in tension could
be observed in the first 8 ms.

Figure 4 – Shock ck simulation of encapsulated lines agasint casing wall; (a) schematics of the impact; (b) Maximum Von Mises on
control lines, for different encapsulation materials.

Tension attenuation occurs differently depending on encapsulation material. Due to the relatively low
impacting velocities in trip-in
in operations,
operations none of results showed risk of yielding in control lines.

A different simulation was conducted considering shock of a clamp agains a casing wall. Figure 5 shows
the maximum Von Mises Tensions observed on control lines situated inside the clamp grooves. grooves On
Figure 5(b),
(b), simulated with legacy encapsulation material, lines show stress beyond yield. Thisindicates
that pre-molded
lded flatpacks tend to concentrate more tension than the plain flatpacks from Figure 4.
6

Figure 5 – Impact simulation for control lines fit inside clamps; (a) simulated assembly; (b) tension peak on control
cont lines, for legacy
encapsulation; (c) tension peak on control lines, for anti-shock
anti encapsulation.

On item (c) of Figure 5,, lines had their tension levels considerably diminuished with the application of
special anti-shock encapsulation.
apsulation. Yield was prevented
prevented with this type of material.

Vibration
Fluid-structure interaction analysees were performed on four different layouts of encapsulation, as shown
inFigure 6.

Figure 6 – Simulated line protections:

tections: (a) B
Bare metal line; (b) Encapsulated
ncapsulated with reinforced PTFE; (c) Encapsulated with anti-shock
anti
material; (d) three-layer

Figure 7 shows the maximum

aximum displacement of a bare metal line, for all the simulated time (2 s).
 7

Figure 7- Maximum displacement (Bare Alloy case, out of scale)

Tension on control lines remained predominantly below 10% of yield stress. Tension concentration was
observed in the inflexion point of Figure 7,, as well as near the connection regions, where encastre
conditions were imposed.Figure 8 shows the result for accumulated fatigue damage, as predicted by eq.
(2).

Figure 8 - Accumulated fatigue damage (Bare Alloy example)

Table 1 summarizes the results encountered for all simulated configurations.

Table 1 – Comparative response of encapsulation layouts to long-term

long vibration

Maximum Displacement Damage Accumulation in Predicted life expectancy

Configuration
(mm) 2 seconds (years)
Bare alloy 8.3 2.38 ∙ 10 1.04
Single-layer reinforced
11.4 2.43 ∙ 10 1.02
PTFE
Single-layer anti-shock
6.9 5.64 ∙ 10 4.40
material
8

Three-layer anti-shock %
9.1 2.54 ∙ 10 9.83
material

All values of life expectancy were low since an extremely severe load case was chosen for the line on
position 1 (Figure 2). Even though this load scenario is non-realistic, it was considered valid for a
comparative analysis between configurations.
Single-layer reinforced PTFE showed no improvement in fatigue when compared to bare lines. Single-
layer anti-shock material could positively contribute to vibration damping and life extension. The three-
layer configuration was the most efficient configuration for fatigue, even though maximum
displacement attenuation was not as pronounced as that observed with the single-layer anti-shock.

Recommendations

Shock mitigation
Viscoelastic materials behave hysteretically when submitted to cyclic tension and relaxation. This
behavior gives to the component the ability to attenuate vibrations through energy dissipation.
Therefore, a damping effect appears that depends significantly on the frequency (&) to which the
material vibrates. The mechanical behavior can be modeled by complex Young’s moodulus, or complex
shear modulus. For instance, the complex shear modulus represented by the equation below can be
broken down into a real (G1), or storage modulus, and an imaginary part (G2), or loss modulus:

' & = '( & + * ∙ '+ & 3

As shown above, there are different G1 and G2 for each frequency of vibration. The loss factor (η), or
structural damping, is defined by eq. (4).

'+ &
, & = 4
'( &
Results show that by increasing the loss factor, higher damping can be achieved.(ASTM-E-756-05,
2017) constains procedures for testing and measuring damping properties of materials. This reference
could be pursued to aid material selection for shock absorbtion.

Vibration mitigation
CFD studies indicated that most of the transversal loading on control lines will take place within a
distance of 0.25 m from flow ports, as shown in Figure 9 (a). It is recommended for lines to be conveyed
inside protective frames throughout this region (Figure 9 (b)).
 9

Figure 9 – Transversal loading on control lines; (a) Forces per unit length in the proximity of flow ports; (b) recommended
encapsulation design

layer protective layout is a known form of mitigating flexural vibration on pipes ( (Hujare
The three-layer
& Shasrabudhe, 2014)). (Kerwin, 1959) was one of the first scientists who suggested this solution.
(Mohammadi & Sedeghati, 2012) concluded that the thickness of the viscoelastic stic layer is fundamental to
increasing dampening. Based on these references and on the results shown on Table 1, the three-layer
layout is recommended around flow control valves. Figure 10 shows a schemati schematic of how this
arrangement should be constructed: two rigid layers (metal lines and outer frame) enclosing a
viscoelastic filling.

Figure 10 –Recommended
Recommended encapsulation schematic near valve flow ports.

Fretting mitigation
Wear can happen in control lines in two ways: firstly, by impact and relative displacement between
control lines and production string. Secondly, wear can happen when lines slide against fixtures used to
constrain them. Valve producers are already aware of the first mode and have been implementing
encapsulation in order to reduce the wear rate between lines and string.
Friction between lines and fixation belts is complicated to address.. According to (Toshio Hattori,
2011), this type of frictionis
is associated with singularity of tension profile
profile,, which causes a tension peak
in the interface of free and restricted areas of the control lines.
ines. This higher tension can lead to crack
generation very quickly, but it will only propagate after after some wear occurs
occur to the contacted zone
changing the compressive tension to crack tip ( (Marco Antonio Meggiolaro, 2016)). 2016) This mechanism
will leadto a late failure, which is hard to predict.
predic
10

For better preventing both types of of fretting/wear mechanisms on control lines, constraing line
displacement seems to be the most effective way of protecting them. As an additional feature, the
encapsulation layout proposed in Figure 10 should be completely fixated onto the valve body.

Conclusion
Industry best practices have recommended that downhole control lines be encapsulated by either
flatpacks or elastomeric sleeves. Even though this type of protection can be satisfying for deployment
operation, it cannot successfully mitigate vibration during production or injection at high rates. In order
to preserve the integrity of control lines, control line producers should design new types of protection,
such as the three-layer structure.Since completion valves are usually unfit to bear complex
encapsulations, some valves would need to have their design rethought.
A novel method of evaluation was proposed for reducing vibration effects on downhole control lines.
Local tension was estimated by thorough replication of the aerodynamical forces during well operation.
Besides estimating the damping factor with regard to displacement, this method allowed for the
calculation of accumulated fatigue damage in the body of the control lines. The most efficient
encapsulation design was found to be the three-layer arrangement. Lines should lie in this type of
protection, especially within a 10-inch range of the flow ports where most of the turbulence occurs.
Fixation of the protecting frame is also advised, so that vibration as well as fretting is even more
mitigated.
For shock phenomena, the influence of the encapsulation design is more straightforward and seems to
depend only on the encapsulation material. Materials with higher loss factor (,) tend to absorb impacts
more efficiently. Lines conveyed inside grooves can be more susceptible to shock damage, probably due
to the presence of residual stresses. Even though grooves offer valuable protection for flatpacks,
conformation stresses can also happen, especially if flatpacks need to be deformedupon installation.

References

3362-AWES-CL-MAIN-FINAL. (2017). AWES Recommended Practice for Control Lines (CL).

3362-AWES-FP-MAIN-FINAL. (2017). AWES Recommended Practice for Flatpack (FP).
3362-AWES-PR-ENCAP. (2020). AWES Recommended Practice for Encapsulation.
Aggogeri, F., Borboni, A., Merlo, A., Pellegrini, N., & Ricatto, R. (2017, March 15). Materials, 10(3),
p. 297. doi:10.3390/ma10030297
ASTM-E-756-05. (2017). Standard Test Method for Measuring Vibration-Damping Properties of
Materials.
Azizian, R., & Torrado, P. (2016). A Practical Approach to Mitigate the Excessive Vibration of a Piping
System Subjected to Flow-Induced Excitation. ASME Pressure Vessel Piping Conference.
doi:10.1115/PVP2016-63088
Bai, J., & Sun, C. (1995). The effect of viscoelastic adhesive layers on structural. Mechanics of
Structures and Machinery, 23(1), pp. 1-16.
Bolding, J. L. (2009). Damaged Control Line Replacement Safety Valve System: Thru-Tubing.
SPE/ICoTA Coiled Tubing & Well Intervention Conference and Exhibition. The Woodlands,
Texas. doi:https://doi.org/10.2118/121407-MS
Cosham, A. H. (2004, Jan). The effect of dents in pipelines - Guidance in the pipeline defect assessment
 11

manual. International Journal of Pressure Vessels and Piping, 81(2), pp. 127-139.
doi:10.1016/j.ijpvp.2003.11.004
Geethamma, V., Asaletha, R., Kalarikkal, N., & Thomas, S. (2014, Oct 09). Vibration and Sound
Damping in Polymers. Resonance, 19, pp. 821-833. doi:https://doi.org/10.1007/s12045-014-
0091-1
Hahn, G., Masson, C., Norris, M. A., & Williams, S. (2015). Subsea Pipeline Vibration Dampers:
Mitigation of Vortex Induced Vibrations. OTC Brasil, . Rio de Janeiro, Brazil.
doi:https://doi.org/10.4043/26169-MS
Hujare, P., & Shasrabudhe, A. (2014). Experimental investigation of damping performance of
viscoelastic material using constrained layer damping treatment. Procedia Materials Science, 5,
pp. 726 – 733.
Kerwin, E. (1959). Damping of Flexural Waves by a Constrained Viscoelastic Layer. The Journal of the
Acoustical Society of America, 31, pp. 952-965. doi:https://doi.org/10.1121/1.1907821
Liman, A. L. (2012, Feb). On the collapse of dented tubes under combined bending and internal
pressure. International Journal of Mechanical Sciences, 55(1), pp. 1-12.
doi:https://doi.org/10.1016/j.ijmecsci.2011.10.005
Marco Antonio Meggiolaro, J. T. (2016). Fatigue Design Techniques: Vol. I - High-Cycle Fatigue: 1.
Rio.
Miner, M. A. (1945). Cumulative damage in fatigue. Journal Applied Mechanics, 12, 159-164.
Min-Kyu Lim, S.-D. O.-Z. (2003). Friction and wear of Inconel 690 and Inconel 600 for steam generator
tube in room temperature water. Nuclear Engineering and Design, 97–105.
Mitchel, R. E. (2010, February 25). Analysis of Control Lines Strapped to Tubing. SPE Drilling and
Completion, 25(02), pp. 210-218. doi:https://doi.org/10.2118/112624-PA
Mohammadi, F., & Sedeghati, R. (2012, Jan). Linear and nonlinear vibration analysis of sandwich
cylindrical shell. International Journal of Mechanical Sciences, 54(1), pp. 156-171.
doi:https://doi.org/10.1016/j.ijmecsci.2011.10.006
Nardi, L. (2018, Sep 19). Caracterização Experimental de Jato Incidente sobre Superfície Côncava em
Espaço Anular. Master Thesis. Rio de Janeiro, Brazil: PUC-Rio.
doi:https://doi.org/10.17771/PUCRio.acad.36406
Rodrigues, D. N. (2017, Oct 24). Intelligent Completions as a Key Technology in Brazilian Presalt. OTC
Brazil 2017. Rio de Janeiro, Brazil. doi:https://doi.org/10.4043/28056-MS
Toshio Hattori, V. T. (2011). Fretting fatigue life estimations based on fretting mechanisms. Tribology
International, 1389–1393.
Zhou, X., Yu, D., Shao, X., Zhang, S., & Wang, S. (2016, Feb). Research and applications of
viscoelastic vibration damping materials: A review. Composite Structures, 136, pp. 460-480.
doi:https://doi.org/10.1016/j.compstruct.2015.10.014